ThreePointEnergy#

class optking.linesearch.ThreePointEnergy(molsys, history, params)[source]#

Bases: LineSearch

Methods Summary

compute_distance()

expected_energy(**kwargs)

Linesearch Algorithms should be able to compute the expected energy based only on the Points.

fit()

Three point parabolic fit.

requires()

reset()

step([fq, energy])

Determine the new step to take in the linesearch procedure.

Methods Documentation

compute_distance()[source]#
expected_energy(**kwargs)[source]#

Linesearch Algorithms should be able to compute the expected energy based only on the Points.

fit()[source]#

Three point parabolic fit. Returns the next point in linesearch.

Returns:

  • step_size (float) – distance to the next point

  • converged (boolean) – True if stepsize is distance to the projected minimum. False if linsearch goes on

requires()[source]#
reset()[source]#
step(fq=None, energy=None, **kwargs)[source]#

Determine the new step to take in the linesearch procedure. (Could be stepping backwards from the previous point).

Parameters:

  • fq (np.ndarary)

  • energy (float)

  • kwargs

Returns:

np.ndarray

Return type:

new step